ebook img

DTIC ADA541077: A Web-Based Library and Algorithm System for Satellite and Airborne Image Products PDF

0.44 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview DTIC ADA541077: A Web-Based Library and Algorithm System for Satellite and Airborne Image Products

DISTRIBUTION STATEMENT A: Approved for public release; distribution is unlimited. A Web-Based Library and Algorithm System for Satellite and Airborne Image Products W. Paul Bissett Florida Environmental Research Institute PO Box 292397 Tampa, FL 33687 USA phone: (800) 928-6402 x102 fax: (800) 928-6402 email: [email protected] Award Number: N0001409C0553 http://www.flenvironmnental.org http://www.onr.navy.mil/sci_tech/32/322/environmental_optics.asp LONG-TERM GOALS The focus of this effort is to create a managed library appliance that supports the Navy METOC personal in their ability to search for in-house hyperspectral and multispectral imagery products, and provide algorithms that create bathymetry estimates from those imagery products on-demand. OBJECTIVES This one-year effort seeks to deliver an image library system that handles terabytes of imagery products with a simple and easy to use index, search, discovery, and delivery functionality. Further, this image library will have an algorithm interface that will allow viewers to “order” an image transformation of their choice, on-demand. We seek to work with a navy customer to place the library system in operational use for satellite and aircraft imagery products. APPROACH The Navy has a myriad of satellite and aircraft imaging systems to support ISR and METOC goals. These systems produce petabytes (PB; 1015 bytes) of imagery and imagery products per year. These imagery products are created worldwide, everyday, with varying time and space resolutions, resulting in thousands of daily products. The selection of any imagery product, or set of imagery products, to support a specific mission is hampered by the inability to quickly and efficiently find the required data. To efficiently find these high-value products, they must be archived in a library system that can rapidly search, discover, and deliver the required information to the intelligence/METOC officer. Once the imagery product is found, it must be delivered to the analyst or operator. However, these imagery and map products may be very large. A single image over a small area (10-100s of square kilometers) may be 10-100s of gigabytes (GB; 109 bytes) in size. Transport of these large imagery products over the internet (or other secure networks) may not be possible because of the bandwidth available between the product library and the analyst. Yet, the information contained in the product still needs to be communicated to the downstream analyst or operator, otherwise the value of the imagery product is lost. While the bandwidth restrictions may be unavoidable, the effective 1 Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 3. DATES COVERED 2010 2. REPORT TYPE 00-00-2010 to 00-00-2010 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER A Web-Based Library and Algorithm System for Satellite and Airborne 5b. GRANT NUMBER Image Products 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Florida Environmental Research Institute,PO Box REPORT NUMBER 292397,Tampa,FL,33687 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES 14. ABSTRACT 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 7 unclassified unclassified unclassified Report (SAR) Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 communication of the required ISR/METOC information may still be possible if the size of the product is reduced, or if the analysis of the product is completed before transport via the network. The reductions in image size, or the server-side rendering of the image product into an intelligence product (e.g. run an algorithm on the image and send only the results) are the most frequent forms of image customization prior to delivery. However, this precludes any form of interaction by the analyst to request changes in the imagery products resulting from changing conditions in the field. This limits the ability of the analyst to be “flexible” in relationship to field or operational conditions, and reduces the overall value of the imagery systems. In other words, if every new request for a slight alteration in the product requires hours or days of work, then the product is less likely to be useful during critical times of operation. To be truly successful, an imagery library system would be combined with the ability to customize the imagery products into high-value derivative maps on-demand. When combined with rapid search capabilities, these customization capabilities would accelerate the time-to- use of high-value imagery products, increasing the value of the products, as well as reducing the risk of ineffective product results. This applied research effort builds upon previous basic and applied research efforts in the development of value-added products from multi- and hyperspectral imaging systems. It specifically seeks to address the issue of search, discovery, and delivery of high-value satellite and aircraft imagery data through internet-enabled systems. It also seeks to enable the on-demand customization of imagery products through the ingestion of specialized image transformations (e.g. algorithms) into the library system. Objectives (1) Deliver an image library system that handles terabytes of imagery products with a simple and easy to use index, search, discovery, and delivery functionality. (2) Develop a transformation and algorithm interface for the library system that will allow viewers to “order” the transformation of their choice, on-demand. (3) Work with a navy customer to place a library system in operational use for satellite and aircraft imagery products. Image Library System The Florida Environmental Research Institute (FERI) has licensed and transitioned its library hyperspectral system for imagery and mapping products to WeoGeo, Inc. FERI works closely with WeoGeo in continuing to develop new capabilities for the library system. In this case, we at FERI will start with the basic implementation of the WeoGeo Library Appliance, and develop the new capabilities required for on-demand transformation and algorithm services. The WeoGeo Library Appliance is a geospatial network appliance specializing in the web-hosting of high bandwidth mapping products. Designed to efficiently search, sort, customize, and deliver an enterprise’s product within seconds of a User’s entry into the system. This system easily handles terabyte (TB) size raster imagery products including color and color-infrared aerial imagery, HyperSpectral Imagery (HSI), LiDAR, and satellite imagery. 2 Figure 1. The WeoGeo Library Appliance is a network appliance that provides full hyperspectral imagery archival and distribution support. The focus of the Library Appliance is to provide quick, efficient access to representations of the full layered, spectral, and spatial resolution data products. This focus allows the Appliance to provide rapid web-based search, sort, and selection capabilities over reduced bandwidth connections. Once a product has been selected, the User is provided with the additional selection options that optimize the product delivery of information layers (e.g. color bands), spatial resolutions, and/or total geographic extent (i.e. geo-clipping). Upon completion of the final product selection, the User’s customized map is rapidly produced, completely independent of human intervention, and placed in a unique FTP folder available only to the user with their individual login credentials. WeoGeo Library’s unique features include: (1) A rolling panel feature that allows for the entire search, sort, selection, and order to be accomplished in the same window. The user never leaves the primary window that was entered. (2) The ability to search from the world to a city within a single viewing point, with data availability graphically given in the highest zoom window. (3) Searching and sorting by key words, locations, provider, projection, data type, map type, etc. (4) A What-You-See-is-What-You-Get (WSYWYG) capability that provides the user/buyer the ability to see exactly the mapping data they are going to order. (5) Direct support for 3-D viewers such as Google Earth and NASA World Wind, so that users may relate these data sets with other sets not available on WeoGeo Server. (6) Direct support of FDGC compliant metadata files. (7) Email notification of request acceptance and data availability. 3 (8) Data set security provided by unique login credential, which allow data sets to be grouped for different client sets and login capabilities. (9) Data request history is maintained for each user for email notification of data product updates, as well as easy reorder of additional products. (10) Batch processing of data sets for easy upload and maintenance with conversions of multiple projection and datums on-the-fly to user specification. (11) Upload of any ortho-rectified raster product in seconds to minutes with live access back to field deployed teams over reduced bandwidth internet connections. Ideal for DoD, DHS, FEMA, or NOAA event response teams deployed in multiple locations following a critical event. Objective 1 – We will deliver a Library Appliance that can inventory, index, search, and deliver Naval-relevant imagery and imagery products. Enabling Transformations and Algorithms The Weogeo Library currently offers limited customization to a select group of raster file types. These file types include JPEG, TIFF/GeoTIFF/BigTIFF, ENVI BIL/BSQ/BIP, ERDAS IMG, and ESRI BIL. Datasets of these types can be requested with the following customizations: geographic clipping/cropping, spatial re-sampling, band/layer selection, output data type format, and output datum and projection. The current interface for making these requests can be seen in Figure 4. All other file types are inventoried and delivered in their native file format. Ideally, customization options should not be limited to a select group of raster formats. Vector data sets, profile measurements, and elevation data (e.g. LiDAR) would all be better served if they could be customized prior to their delivery. Each of these data sets types would have their own customization options and associated metadata. For example, CTD profiles have a geographic component, and it would be desirable to be able to subsection the profile by time or depth prior to delivery of the profile to an end-users. Furthermore, user demanded customization of the data products might be more than just simple subsections of the data. These requests could be more complex, such as to run an algorithm like a spectrum matching approach to retrieve Inherent Optical Properties (IOPs) from hyperspectral imagery. These types of data transformations increase the value of the imagery product, while at the same time reducing the data volume and bandwidth requirements for transport via the internet. The current means of handling the limited data customization may be seen in Figure 2. The viewing and selection operations seen in Figure 2 will need to be redesigned and replaced in order to make it flexible enough to handle other data transformations and algorithms. FERI will design the Naval- relevant template for WeoGeo, with a specific implementation of the spectrum matching approach to inverting hyperspectral imagery created by Drs. C. Mobley (Sequoia Scientific) and P. Bissett (FERI). 4 Objective 2 – We will design and create the ability to transform imagery products via the Library Appliance, with the specific use of the spectrum matching algorithms developed by Sequoia Scientific and FERI. Testing and Implementation of Library This project will result in the delivery of a WeoGeo Library with algorithm capabilities to an operational Naval partner. Our target partner is the METOC Sensing Programs Transitions & Integration at the Naval Meteorology and Oceanography Command (NAVO). We will try to work closely with T. Bowers, Commander of that operational group. We will also work closely with Naval Research Laboratory partners at the Stennis Space Center and in Washington, DC to provide pathways for their algorithms to be incorporated into the NAVO WeoGeo Library. Objective 3 – We will deliver of the augmented Library Appliance to Navy transition partner for testing and implementation within an operational environmental. Figure 2. Current customization screen for imagery products on WeoGeo Library Appliance. The data Preview Panel (panel 4 of the data selection process) provides basic metadata and access to enhanced descriptions of the data product. The Customize Panel (panel 5 of the data selection process) provides the data customization parameters for spatial resampling, band selection, file format, data projection, and geospatial clipping 5 WORK COMPLETED Objectives 1 and 2 have been completed. Figures 3 and 4 show the web-page parameter selection screens for the kNN algorithm. Figure 3. Parameter selection screen for the kNN algorithm on the ONR WeoGeo Library Appliance. The screen allows the user to select the wavelengths to be used to for the spectrum matching, matching result, depth limit, number of neighbors, and statistical metrics to determine best match between the database spectrum and the image spectrum. Objective 3 has not been completed. Our original contact at NAVO has moved from his post, and we have found no one at NAVO who is interested in becoming the accepting agent for this Library Appliance. We continue to seek someone within a Naval command who would be an appropriate recipient for this system. 6 Figure 4. Parameter selection screen for the kNN algorithm on the ONR WeoGeo Library Appliance. This figure shows how drop down menus are used to select the parameters. In this example, the statistical metric to be used to match the spectrum will be Manhattan distance calculations between the database spectrum and the image spectrum. IMPACT/APPLICATIONS This effort will deliver a Library Appliance that will produce bathymetry estimation from multispectral and hyperspectral imagery products in Very Shallow Waters (VSW). This will support Navy METOC personal in their ability to search for in-house hyperspectral and multispectral imagery products, and warfigher support imagery products on-demand. RELATED PROJECTS This work is transitioning VSW algorithms developed by Dr. Curtis D. Mobley at Sequoia Scientific, Inc., and Dr. Paul Bissett at FERI, under other 6.1/6.2 program funding. 7

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.